Every Attention Matters: An Efficient Hybrid Architecture for Long-Context Reasoning Paper • 2510.19338 • Published Oct 22 • 114
Fast Chain-of-Thought: A Glance of Future from Parallel Decoding Leads to Answers Faster Paper • 2311.08263 • Published Nov 14, 2023 • 16
MultiLoRA: Democratizing LoRA for Better Multi-Task Learning Paper • 2311.11501 • Published Nov 20, 2023 • 37